The U.S. Environmental Protection Agency’s (EPA) Water Infrastructure Resiliency and Finance Center, in collaboration with the...
New test improves prediction of system and media performance
The most common question from customers of water treatment systems utilizing adsorbent media is, “How long will the media/system perform?” The most common answer may be a number of months or years but should be more elaborative than a mere number. The most honest answer is, “It depends on the water chemistry.”
Even after a thorough review of water chemistry, estimating system/media performance can be difficult. Periodic monitoring is recommended regardless of estimates, but accurate media or system performance estimates can be made using a variety
Traditional Performance Testing Methods
An onsite pilot is going to be the most accurate predictor as long as the needed time, equipment and trained personnel are available to conduct and monitor this testing. When an onsite pilot is not practical, customers often rely on the advice of the supplier or treatment media manufacturer to predict performance. Suppliers and manufacturers have many tools available to develop these performance estimates. Empirical data from existing operating systems, prior field trials and laboratory tests are assessed and a variety of laboratory testing methods are employed to evaluate specific water chemistry. These data are used to develop performance-indicating models based on multi-variable water chemistries. As databases of water chemistries and media/system operating performances grow, the resulting predictive model becomes increasingly more accurate.
Laboratory-scale evaluations for media performance in water from a specific site can be conducted utilizing static or dynamic testing methods. Static methods like batch equilibrium (jar) testing and adsorption isotherms can be performed in less than a day and can give an accurate indication of the total media capacity. Unfortunately, media performance can be strongly influenced by:
These factors make these static tests more applicable to the media’s affinity for a specific contaminant than predicting operational run length.
Dynamic laboratory testing can either replicate field operating conditions or be accelerated. Replicating field operating conditions, like empty bed contact time, requires a large volume of water from the test site to be shipped to the testing laboratory. Even if a scaled down media bed volume of 200 mL was assumed capable of processing 15,000 bed volumes, more than 700 gal of challenge water would be required and testing could take upwards of a month to complete.
The need to have more timely, accurate results and the desire to avoid shipping large volumes of water has resulted in the widespread usage of accelerated testing like the rapid small-scale column test (RSSCT). RSSCT has proven to give results comparable to field performance and is a widely accepted method. Performing RSSCT requires the test media to be “ground” and screened to a specific particle size, which is not always advisable for some medias for which grinding alters the intended structure (e.g. impregnated resins) or physical strength. Graver Technologies LLC has begun to see success utilizing an alternative dynamic test, the accelerated segmented column test (ASCT) for estimating performance of its nano-titanium oxide arsenic adsorbent media, MetSorb HMRG.
Another Tool in the Tool Belt
The ASCT is based on a desorption model for organic contaminants on granular activated carbon. The test follows the contaminant “wave front” through the media bed in segmented fractions. Each segmented fraction represents a time interval (15 seconds, 30 seconds, 60 seconds, etc.) of the full-scale media bed. The volume at which the effluent from each column segment reaches the U.S. Environmental Protection Agency maximum contaminant level of 10 parts per billion (ppb) endpoint is used in estimating the total volume of a specific water that can be treated by the MetSorb HMRG arsenic reduction media.
The laboratory ASCT evaluation can be performed in less than two weeks and utilizes as little as 120 gal of challenge water. The successful parallels between pilot and lab performance are promoting further development of this media life estimation tool to support cost-effective, predictive methods of determining media/system operating life.